Change search
ReferencesLink to record
Permanent link

Direct link
Positioning techniques alternative to GPS
2003 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Ericsson AB has a wide range of tools in the TEMS portfolio with the purpose to measure speech quality, data throughput, coverage etc of mobile networks. These tools require some method for positioning the measurements on a map. This is normally performed by GPS, but due to the lack of GPS coverage in certain areas, another positioning technique must be adopted. The task of the Master’s thesis was to investigate positioning techniques alternative to GPS, with a focus on positioning using accelerometers. The proposed system uses an approach where the distance travelled is computed by counting the steps of the user. Steps are detected using a step detection algorithm on data deduced from a sensor module placed on the lower back of the user. The sensor contains a 2-axis accelerometer with one axis vertical of the user and one axis in the direction of the walk. In the prototype the sensor is connected to a laptop carried by the user. In a final version, however, the system may be carried in a backpack, improving usability substantially. The length of each step is determined by a step model, which uses the fact that step frequency and step length usually are well correlated. Due to the nonlinear nature of human walking, the step model is calibrated on-line using an optimal estimator. The direction of the displacement can be calculated using gyroscopes and electronic compasses, but since the hardware for such a system either must be constructed or purchased at a very high cost, a different approach is adopted. The system uses a predefined route, which consists of waypoints defined by the user. When the user reaches a waypoint the user indicates it by pressing a button on the laptop. The actual position is updated and the step model is recalibrated using new data. Experimental results using the prototype have shown reliable performance. Tests conducted at a 400-m running track resulted in calculated distances with an error of less than two percent.

Place, publisher, year, edition, pages
Keyword [en]
Technology, optimal filter, kalman filter, step counting, positioning, GPS
Keyword [sv]
URN: urn:nbn:se:ltu:diva-50769ISRN: LTU-EX--03/003--SELocal ID: 7febf30b-f38b-4b14-9bba-305788d60f51OAI: diva2:1024131
Subject / course
Student thesis, at least 30 credits
Educational program
Computer Science and Engineering, master's level
Validerat; 20101217 (root)Available from: 2016-10-04 Created: 2016-10-04Bibliographically approved

Open Access in DiVA

No full text

Search outside of DiVA

GoogleGoogle Scholar

Total: 6 hits
ReferencesLink to record
Permanent link

Direct link